Objective Pregnancy has a well documented effect on relapse risk in multiple sclerosis (MS). Prospective studies have reported a significant decline by two-thirds in the rate of relapses during the third trimester of pregnancy and a significant increase by two-thirds during the first 3 months postpartum. However, it is unclear as to whether there are any long term effects on disability.
Methods Data were collated from clinical records and family histories systematically collected from the University of British Columbia MS Clinic.
Results Clinical and term pregnancy data were available from 2105 female MS patients. MS patients having children after MS onset took the longest time to reach an Expanded Disability Status Scale (EDSS) score of 6 (mean 22.9 years) and patients having children before MS onset were the quickest (mean 13.2 years). However, these effects were not related to term pregnancy and were fully accounted for by age of MS onset.
Conclusions Pregnancy had no effect on the time to reach an EDSS score 6. As MS predominantly affects women of childbearing age, women with MS can be reassured that term pregnancies do not appear to have any long term effects on disability.
Statistics from Altmetric.com
Multiple sclerosis (MS) is an inflammatory disease of the CNS.1 MS predominantly affects women of childbearing age1 and therefore any effect of pregnancy itself on MS is a key issue for patients. Many physiological changes occur in pregnancy, beginning very early in gestation.2 These natural changes have the potential to interact with disease related factors in women with MS and thus precipitate complications with respect to the disease course.
Results from the Pregnancy in Multiple Sclerosis (PRIMS) study showed that pregnancy is probably neutral overall in terms of MS disease activity. The PRIMS study showed an approximate 70% reduction in relapse rate during the third trimester of gestation and a similar increase above background pre-pregnancy relapse rates for the first 3 months postpartum.3
Studies regarding the effects of term pregnancies on long term disability are however conflicting. Early studies could find no long term effect of term pregnancies on MS course4–6 but more recent studies have suggested a beneficial effect of term pregnancies on MS,7 ,8 with female patients having children taking longer to reach disability milestones, as measured by the Expanded Disability Status Scale (EDSS),9 than female patients who did not. Discrepancies between studies may result from relatively small sample sizes, ascertainment of non-population based cases and varying classifications of pregnancy and clinical outcome. In this study, we used data from the large longitudinal population based databases from the University of British Columbia MS Clinic (UBC Clinic) to investigate the long term effects of term pregnancies on MS natural history.
Standard protocol approvals, registrations and patient consent
Data collection and de-identified analyses were in accordance with UBC Clinic patient consents, approved by UBC's Clinical Research Ethics Board.
This study was limited to female MS patients who delivered full term live births (hereafter referred to as ‘births’) to minimise confounding factors from stillbirths, prematurity and pregnancy loss (spontaneous and induced). A pregnancy that has reached 37–40 weeks is considered full term as all organs and systems are fully developed by 37 weeks and the infant born at this time will be ready for life outside of the womb (http://en.mimi.hu/pregnancy/full_term.html).
Data on births were serially collected from consecutive patients attending the UBC Clinic. Since its inception in September 1980, all attendees have had detailed family history information, including pregnancy history, systematically collected by genetic counsellors. Data were validated by checking medical records as necessary, with appropriate consent. UBC Clinic patients have been followed longitudinally as per the clinic mandate. MS was diagnosed using the diagnostic criteria of Poser et al10 and/or McDonald et al.11 Note that individuals seen prior to the adaptation of the Poser et al criteria were re-evaluated on follow-up and only included here if they then fulfilled the criteria. Changes in family (births, deaths, new cases of MS or other major illnesses) were reported and family histories updated as appropriate.
For the purpose of this study, the study cohort included female MS patients with information on year at MS onset, EDSS scores up to 6 and complete ‘term pregnancy status’—(1) whether or not they had any live births (“yes” or “no”) and (2) if “yes”, the year of birth of each child known. To the best of our knowledge, there was no selection bias. All consecutive women with completed data information were included.
Year of MS onset is the year when the initial clinical symptom(s) was observed, as recorded on the UBC Clinic medical record. This is the ‘best estimate’ and often determined retrospectively at the initial UBC Clinic visit through discussion with the patient and healthcare provider (physician, nurse) together with a review of the medical records provided on patient referral or by request. For analyses as necessary, ‘age’ at onset was calculated using year of birth and year of onset.
EDSS is measured by clinic nurses and/or clinic neurologists. Most patients are seen on an annual basis or more often if they have MS relapses and/or any medical issues. We used an EDSS score of 6 in our outcome analyses as this is a robust and clinically relevant disability milestone.12 To be rated as ‘EDSS 6’, a patient had to reach an EDSS score of 6 in one assessment and EDSS score ≥6 in the next assessment, at least 3 months later, to ensure that the original score did not reflect a relapse.
Four covariables—‘term pregnancy status’ (four categories), year of birth (four categories), ‘age’ at MS onset (seven categories) and MS course (two categories)—were considered here. The joint relationship of the four covariates on the time to EDSS 6 was examined using Cox proportional hazards analysis. Spearman correlation was used to investigate the relationship between age of onset and having children.
A cohort of 3114 female MS patients with EDSS scores and pregnancy history information was identified from the UBC Clinic database. Of this cohort, 2936 patients had information on year at MS onset and complete ‘term pregnancy status’.
Of the 2936 women first identified for inclusion, 831 were excluded for the following reasons:
A patient reached EDSS 6 only once and there had not been another assessment at the time of data extraction—it was not possible to schedule this just for the purpose of this study.
A patient reached EDSS 6 but at the next assessment the score was <5.5.
The initial recorded EDSS score for a patient was >6.0 and prior records with EDSS scores could not be obtained; there was no way of knowing when EDSS=6.0 was initially reached.
Thus the final study cohort consisted of 2105 unique female MS patients (table 1). Mean age of the entire female patient group was 47.7 years (SD 11.7), mean EDSS score was 3.2 (SD 2.1) and mean disease duration was 15.2 years (SD 10.5) at the last evaluation. The average number of clinic visits was 9.0 (SD 6.9), the average number of EDSS measurements was 5.7 (SD=5.2) and the average number of follow-up years was 8.6 (SD 6.9). Thus the study cohort averaged about one EDSS measurement per year.
Most patients (43.3%) had children before MS onset, slightly less had no children (34.0%) and even fewer had children after MS onset (13.8%). About 5% had children both before and after MS onset (4.6%) and 4.4% had the initial MS onset during a pregnancy (symptoms during the gestational period).
In the subsequent analyses, 93 patients who had their MS onset during the pregnancy year were excluded. Only MS patients with no children (n=715), MS patients who had children before MS onset and no children at any other time (n=911), MS patients who had children after MS onset and no children at any other time (n=290) and MS patients who had children both before and after MS onset and not during MS onset (n=96) were included. To assess the effects of childbirth on disability, we used time to reach EDSS 6. This was longest (mean 22.9 years) for women having children after MS onset and shortest for those having children before MS onset (mean 13.2 years). The Spearman correlation between age of onset and the time to EDSS 6 (370 cases with EDSS 6) was r=−0.51, p<2.2×10−10.
Data on MS course at onset (primary progressive or relapsing remitting) was available for 1888 patients. The joint relationship of the four covariates (‘term pregnancy status’, birth cohort, ‘age’ at MS onset and MS course) on the time to reach EDSS 6 was examined using Cox proportional hazards analysis. ‘Term pregnancy status’ of the female MS patients was not significant and thus was subsequently dropped as a covariate for the analyses. The significant covariates for time to EDSS 6 were MS course, ‘age’ at MS onset and year of birth of patients. The different combinations of interaction terms of these three covariates were not significant. HRs are given in table 2 for the three significant covariates.
The preponderance of women of childbearing age affected with MS makes it essential to understand the effects of term live births on the course of the disease to reliably educate patients and their healthcare professionals. In the present study, we observed no effect of term pregnancies on the time to reach EDSS 6, a robust and clinically relevant disability milestone at which a walking aid is required. While there were differences in the time taken to reach EDSS 6 between mothers who had children before MS onset compared with mothers who had children after MS onset, this effect was entirely accounted for by the age of MS onset, a characteristic well known to affect clinical outcomes.13 The literature is conflicting with regard to this question, but our study is in agreement with the majority of studies which could not find any effect of term pregnancies on long term disability.4–6 ,14 Two studies did find a beneficial effect of having children after MS onset7 ,8 but it is impossible to exclude conceptive behaviour bias (ie, patients with a mild course would be more likely to have children) as explaining these results.
It is of interest that women who had children before MS onset had a significantly greater age of MS onset than other pregnancy groups, which has also been documented by others.8 ,15 It has been suggested that term pregnancies before MS onset, at least partially, might have influenced the age at MS onset through the effects of oestrogens on the immune system.15 This intriguing possibility needs to be tested.
There are limitations to our study. Despite having the largest cohort to date to address this topic, the number of patients who reached EDSS 6 at the time of this study was modest. We also did not have information on relapse frequency or MRI measures. Furthermore, data on the use of immunomodulatory treatment, which may play a role in the time taken to reach EDSS 6,16–18 were unavailable. However, when investigating individuals who had no clinical relapses (primary progressive MS patients) versus those with more than one (relapsing remitting MS patients), there was still no effect observed of pregnancy on time to EDSS 6. We also did not have details on pregnancies that may have ended in abortion, miscarriage or stillbirth.
In conclusion, women with MS should be reassured that a term live birth has no long term effects on disease course.
The authors thank all patients who generously participated in this study.
Funding This work was funded by the Multiple Sclerosis Society of Canada Scientific Research Foundation and the Wellcome Trust (Grant No 075491/Z/04).
Competing interests None.
Ethics approval Ethics approval was provided by the University of British Columbia.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.